The goals for the Time Capsule File System (TCFS) format, as specified in appendix A, include the necessary provisions for platform and medium independence. It can even be used to encode non-traditional files with multiple data segments - for instance, files from the Apple Macintosh Hierarchical File System. However, TCFS does not stop there, because we know neither what computers nor what data storage will look like in the future. With TCFS, future users will easily be able to reconstruct the format from the actual archives if all the TCFS specification documents are lost.
We made as few assumptions as possible about the archive file's reader, but we were forced to make some assumptions in the format since it would have been extraordinarily difficult to assume zero knowledge. TCFS first assumes that the reader of the file has an understanding of English (or can still find someone who does understand English), since most of the content of the files is worthless if you were to decode it and not understand the language in which it is expressed. Future archivists must also have a knowledge of the ASCII character set to decode the file headers, but the body of the file has no such restrictions. (A file encoded in the EBCDIC character set would have the headers encoded in ASCII, while the data could still be in some variety of EBCDIC.) Finally, one must understand the concept of a file as a metaphor for organizing information. This is a particularly important assumption. Without this idea, the archived information is mostly gibberish, and at the rapid rate computers and software are evolving, the idea of a file may be abandoned quickly for something more versatile, like some variety of persistent objects.